Noncommutative K-correspondence categories, simplicial sets and pro C -algebras
|
|
- Olivia Booker
- 6 years ago
- Views:
Transcription
1 Noncommutative K-correspondence categories, simplicial sets and pro C -algebras Snigdhayan MAHANTA Institut des Hautes Études Scientifiques 35, route de Chartres Bures-sur-Yvette (France) Août 2009 IHES/M/09/40
2 NONCOMMUTATIVE K-CORRESPONDENCE CATEGORIES, SIMPLICIAL SETS AND PRO C -ALGEBRAS SNIGDHAYAN MAHANTA Abstract. We construct an additive functor from the category of separable C -algebras with morphisms enriched over Kasparov s KK 0 -groups to the category of noncommutative K-correspondences NCC K dg, whose objects are small DG categories and morphisms are given by the equivalence classes of some DG bimodules up to a certain K-theoretic identification. Motivated by a construction of Cuntz we associate a pro C -algebra to any simplicial set, which is functorial with respect to proper maps of simplicial sets and those of pro C - algebras. This construction respects homotopy between proper maps after enforcing matrix stability on the category of pro C -algebras. The first result can be used to deduce derived Morita equivalence between DG categories of topological bundles associated to separable C - algebras up to a K-theoretic identification from the knowledge of KK-equivalence between the C -algebras. The second construction gives an indication that one can possibly develop a noncommutative proper homotopy theory in the context of topological algebras, e.g., pro C -algebras. Introduction Experts believe that in any category of noncommutative spaces correspondence-like morphisms should also be included, e.g., motivated morphisms in [42]. Such morphisms are given by some bimodules or generalizations thereof that induce well-defined morphisms on most (co)homological constructions that we know. Such correspondences can be naturally seen as morphisms over some desirable enrichments of the category of noncommutative spaces, for instance, additive or spectral ones. Furthermore, most of the interesting (co)homological invariants that we know should factor through this category. The philosophy closely adheres to that of motivic homotopy theory [26] but including noncommutative objects. Bearing in mind the relative importance of K-theory, we study certain categories, which we call noncommutative K-correspondence categories, whose morphisms naturally include those between topological K-theories. More precisely as a category of noncommutative K-correspondences in the operator algebraic setting we consider the well-known category KK C with Kasparov s bivariant K-theory elements as morphisms, which, in fact, has a triangulated structure [45]. This category seems to have the good formal properties to be considered as an absolute correspondence category, at least in the context of separable C - algebras. Indeed, this category appears in the context of noncommutative motives in [15]. The additivized Morita homotopy category of DG (differential graded) categories Hmo 0, whose construction was outlined in [62], is potentially a good candidate for the absolute category of noncommutative correspondences in the setting of DG categories. Kontsevich calls the same category as the category of noncommutative motives in [40]. All additive invariants of DG categories factor through this category [62]. However, for the applications that we have in mind a more decisive localization is needed. In certain applications to (co)homological duality statements in physics one would like to construct morphisms which induce isomorphisms 1
3 between topological K-theories (or other generalized cohomology theories). Therefore, we invert morphisms which induce homotopy equivalences between Waldhausen s K-theory spectra [63] of DG categories and further perform a group completion of the morphisms. This construction is admittedly not entirely satisfactory as the author is unaware of good method of inverting a class of maps in a combinatorial model category (see however [2]). We also show how this category is related to the topological K-theory of separable C -algebras. It is plausible that some other localization will be beneficial for a different application. From our point of view, one can obtain duality results fairly easily if one is less ambitious, i.e., instead of trying to prove absolute equivalence of DG or A -categories, if one tries to prove them after a suitable localization one can be more successful. In the first section we describe the noncommutative DG K-correspondence category NCC K dg and the noncommutative C -K-correspondence category KK C. Then we construct our additive functor Top K fib : KK C NCCK dg (see Theorem 1.16). The functor Top K fib ( ) roughly sends a unital C -algebra to its bounded DG category of complexes of vector bundles (or finitely generated projective modules). At the heart of this construction lies Quillen s description of the K 0 -group of nonunital algebras [52], which is particularly well behaved for C -algebras. The interest lies mostly in isomorphisms and by functoriality this result implies that an isomorphism in KK C would translate to an isomorphism of DG categories in NCC K dg. It is clear that isomorphisms in NCC K dg are weaker than those in Hmo 0. In the classification programme of C -algebras the Kirchberg Phillips Theorem states that two stable Kirchberg algebras are KK-equivalent if and only if they are -isomorphic [39, 50] and therefore KK-equivalent stable Kirchberg algebras will have isomorphic DG categories of topological bundles, whose isomorphisms will be more tractable. This result is probably by itself not very interesting as it is purely a topological statement. However, by incorporating more structures into a C -algebra (for example, a curved topological DGA) one can hopefully produce more instances of noncommutative dualities as in [4, 5, 6] simply from the well studied KK-isomorphisms. Purely in the operator algebraic context the connection between KK-dualities and noncommutative T-dualities have been explored in [9, 10]. The functor Top K fib seems to be related to some nontrivial isomorphisms like the Fourier Mukai type dualities for tori (see Example 1.20). One can also deduce that for any nuclear separable C -algebra the DG category Top K fib ( ) is invariant in NCC K dg under strong deformation if and its strong deformation are suitably homotopy equivalent (see subsection 1.6). Similar results at the level of K-theory groups already exist in the literature [22] (see also [56] for a survey). We also discuss the concept of homological T-dualities and its connection with that of topological C -K-correspondences briefly (see subsection 1.7). It would be interesting to extend our construction of Top K fib ( ) to the context of purely algebraic bivariant K-theory developed by Cortiñas Thom [18], at least in the H-unital case. Furthermore, it is also plausible that our additive functor could be promoted to an exact functor between triangulated categories, by localizing DGcat along the nonconnective K-theory spectrum (see [63]) The second section is inspired by some emergent connections between homotopy theory and noncommutative topology. Several problems in noncommutative topology, including the Baum Connes conjecture, have some level of properness built into them. From our point of view the objects themselves need not be finite (or compact) but the maps interconnecting them should be proper in some suitable manner and compose well to form a category. Cuntz constructed a universal noncommutative C -algebra using generators and relations from a 2
4 locally finite simplicial complex to give a conceptual understanding of the Baum Connes assembly map in [20]. We associate a noncommutative pro C -algebra to any simplicial set (without any finiteness assumption), which is functorial with respect to proper maps of simplicial sets (see Definition 2.7). Then we show that our construction induces a functor between the category of simplicial sets with, what we call, proper homotopy classes of proper maps between them and the matrix stabilized category of pro C -algebras with homotopy classes of proper or nondegenerate maps between them. Although we mostly deal with the category of pro C -algebras with proper -homomorphisms between them, in order to enforce matrix stability we enlarge the morphisms by corner embeddings, which are not proper, and invert these maps formally. However, the formal inverses of the corner embeddings are in some sense proper. Another sticking point is that we cannot ensure that our construction produces continuous -homomorphisms between pro C -algebras. Therefore we work with arbitrary -homomorphisms which become automatically continuous if the domain pro C - algebra is actually a σ-c -algebra or a countable inverse limit of C -algebras. It seems plausible that without enlarging the category of C -algebras to include certain limits of topological -algebras, it is not possible to construct a Quillen model category structure on it. After appropriate enlargement (containing pro C -algebras) there does exist a cofibrantly generated model category structure with KK -equivalences (or K -equivalences) as weak equivalences [33]. Our result might suggest that the matrix stabilized category of pro C - algebras with proper maps between them is amenable to noncommutative proper (or infinite) homotopy theory as explained in, e.g., [3]. The author recently learnt that a very general framework of homotopy theory in the context of C -algebras has been developed by Østvær [47]. Although the two sections presented in this article seem unrelated the author hopes to make the connection clearer in future. A general remark is in order here. A C -algebra or an abstract DG category is not very geometric in nature. The appropriate objects for geometry should be something close to Connes spectral triples (see, e.g., the reconstruction Theorem [13]) and presumably a pretriangulated DG category whose homotopy category is geometric in the sense of Kontsevich [41]. Therefore, our results should be viewed in the realm of noncommutative topology on which interesting geometric structures can be built. In particular, our K-correspondence categories should be regarded as purely topological (not in the sense of a topological enrichment) categories. In order to make the constructions a bit more sensitive to the norm structures on the topological algebras, one might consider replacing the DG category of C-linear spaces (noncommutative point) throughout by the DG category associated to the category of locally convex topological vector spaces, which admits the structure of a quasiabelian category [58, 51]. Notations and conventions: We do not assume our algebras to be commutative or unital unless explicitly stated so. In Section 1 we require our C -algebras to be separable, which from the point of view of topology requires the spaces to be metrizable. Many constructions in geometry require paracompactness (or an argument involving partition of unity) and from that perspective the separability assumption is quite natural. Moreover, the technical issues of KK-theory are properly understood only in the separable case. The focus of Section 2 is combinatorial topology and hence we do away with the separability assumption and in fact we work with a larger category of pro C -algebras or inverse limit C -algebras. We are going 3
5 to work over a ground field k and while working with C -algebras the field k will be tacitly assumed to be C. Unless otherwise stated, all functors are also assumed to be covariant and appropriately derived (whenever necessary), although we shall use the underived notation for brevity, e.g., we shall write for L. The tensor products of C -algebras are suitably completed and since in all cases one of the algebras is nuclear we do not need to worry about the distinction between maximal and minimal tensor products. Throughout this article we make use of the language of model categories and simplicial homotopy theory whose details we have left out. The standard references for them are [53, 32, 30, 27]. Acknowledgements. The author is extremely grateful to P. Goerss, B. Keller, Matilde Marcolli, R. Meyer, Fernando Muro, J. Rosenberg, G. Tabuada and B. Toën for several correspondences answering various questions. The author would like to thank Benoît Jacob, S. Krishnan and M. Schlichting for some helpful discussions. The author is particularly indebted to A. Connes, G. Elliott and R. Meyer for pointing out several inaccuracies in an initial draft of the article. The mistakes that might remain are solely the author s responsibility. The author also gratefully acknowledges the hospitality of the department of mathematics at the University of Toronto, Fields Institute and Institut des Hautes Études Scientifiques, where much of this work was carried out. 1. Noncommutative K-correspondence categories The general philosophy in this paradigm is that an object can be studied by its category of representations. This category of representations could be endowed with an abelian, triangulated, differential graded (DG) or some symmetric monoidal structure. It appears that dealing with abelian or triangulated categories is deficient from the homotopy theoretic point of view. It is better to work with the entire RHom complex of morphisms, which retains cochain level information rather than simply the zeroth cohomology group as in triangulated categories. Hence, from our perspective the appropriate category structure is that of a category enriched over cochain complexes, i.e., a DG category. It is often convenient to localize along certain morphisms for various purposes. For (co)homological constructions in geometry localizing along quasi-equivalences seems quite natural. This leads us to Keller s construction of the derived category of a DG category, which is essentially the category of modules (or representations) of the DG category up to homotopy. This derived category is itself is a triangulated category and for most practical purposes it suffices to work with this triangulated category of modules over the DG category or some suitable triangulated subcategory thereof. We recall some basic facts about DG categories and noncommutative geometry below. There is some freedom in choosing the way one would like to represent the known constructions in geometry in this setting. Let us reiterate that the noncommutative K-correspondence categories introduced below should be viewed simply as topological models The category of small DG categories DGcat. The basic references for the background material, that we require, about the category of small DG categories are [37] and [62]. In this setting the noncommutative spaces are viewed as small DG categories, i.e., categories enriched over the symmetric monoidal category of cochain complexes of k-linear spaces. Let DGcat stand for the category of all small DG categories. The morphisms in this category are DG functors, i.e., (enriched) functors inducing morphisms of Hom-complexes. Henceforth, 4
6 unless otherwise stated, all DG categories will be small. We provide one generic example of a class of DG categories which will be useful for later purposes. Example 1.1. Given any k-linear category M it is possible to construct a DG category C dg (M) with cochain complexes (M, d M ) over M as objects and setting Hom(M, N ) = n Hom(M, N ) n, where Hom(M, N ) n denotes the component of morphisms of degree n, i.e., f n : M N [n] and whose differential is the graded commutator d M f n ( 1) n f n d N. It is easily seen that the zeroth cocycle category Z 0 (C dg (M)) reduces to the category of cochain complexes over M and the zeroth cohomology category H 0 (C dg (M)) produces the homotopy category of complexes over M. Thus, given any DG category C the category H 0 (C) is called the homotopy category of C. If the objects of C dg (M) are taken to be chain (instead of cochain) complexes over M one needs to set the n-th graded component of the morphism Hom(M, N ) n = Hom(M, N [ n]). Now we recall the notion of the derived category of a DG category as in [36]. Let D be a small DG category. A right DG D-module is by definition a DG functor M : D op C dg (k), where C dg (k) denotes the DG category of cochain complexes of k-linear spaces. We denote the DG category of right DG modules over D by D dg (D). It generalizes the notion of a right module over an associative unital k-algebra A. Indeed, viewing A as a category with one object such that End( ) = A, a functor from the oposite category to k-linear spaces is just a k-linear space M (the image of ) with a k-algebra homomorphism A op End k (M) making M a right A-module. Every object X of D defines canonically what is called a free or representable right D-module X := Hom D (, X). A morphism of DG modules f : L M is by definition a morphism (natural transformation) of DG functors such that fx : LX MX is a morphism of complexes for all X Obj(D). We call such an f a quasi-isomorphism if f X is a quasi-isomorphism for all X, i.e., f X induces isomorphism on cohomologies. The derived category D(D) of D is defined to be the localization of the category D dg (D) with respect to the class of quasi-isomorphisms. The category D(D) attains a triangulated structure with the translation induced by the shift of complexes and triangles coming from short exact sequence of complexes. The Yoneda functor X X induces an embedding H 0 (D) D(D). Definition 1.2. The triangulated subcategory of D(D) generated by the free DG D-modules X under translations in both directions, extensions and passage to direct factors is called the perfect derived category and denoted by per(d). Its objects are called perfect modules. A DG category D is said to be pretriangulated if the above-mentioned Yoneda functor induces an equivalence H 0 (D) per(d). Remark 1.3. The homotopy category of a pretriangulated category has a triangulated category structure which is idempotent complete. There is a canonical DG version, denoted by per dg (D), whose homotopy category is per(d). The construction is analogous to that of Example 1.1. Considering an associative unital algebra A as a DG category one finds that per(a) is equivalent to the homotopy category of bounded complexes of finitely generated projective modules over A The Morita model structure on DGcat. A DG functor F : C D is called a Morita morphism if it induces an exact equivalence F : D(D) D(C). There are many equivalent formulations of a Morita morphism. The one that we have chosen is perhaps the most direct 5
7 generalization to the derived setting of (algebraic) Morita morphism as an equivalence of module categories. Thanks to Tabuada [62] we know that DGcat has a cofibrantly generated Quillen model category structure, where the weak equivalences are the Morita morphisms and the fibrant objects are pretriangulated DG categories. The category of noncommutative spaces NCS dg is defined to be the localization of DGcat with respect to the Morita morphisms, i.e., the Morita homotopy category of DGcat. One should bear in mind that a Morita morphism is, in general, weaker than what is called a quasi-equivalence, which generalizes the concept of a quasi-isomorphism. Given any DG category A one constructs per dg (A) as its pretriangulated replacement or fibrant replacement. Using the fibrant replacement functor it is possible to view NCS dg as a full subcategory of DGcat localized along quasi-equivalences consisting of pretriangulated DG categories. The category NCS dg has an inner Hom functor which we denote by rep dg (,?) [64]. For the benefit of the reader we recall briefly its construction. Given any DG category A one can construct a C dg (k)-enriched model category structure on the DG category of right DG A- modules D dg (A), whose homotopy category turns out to be equivalent to the derived category of A [64]. Let Int(A) denote the category of cofibrant-fibrant objects of this model category, which may be regarded as a C dg (k)-enrichment of the derived category of A. If C and D are DG categories then their inner DG category rep dg (C, D) is by definition Int(D dg (D op C)). A DG functor φ : C D naturally gives rise to a D op C-module M φ, i.e., M φ (?) = Hom D (φ(?), ). This is one advantage of working in the DG setting, i.e., every morphism (not necessarily isomorphisms) in NCS dg becomes a generalized DG bimodule morphism or a noncommutative correspondence. In the geometric triangulated setting, e.g., when the triangulated category is of the form D b (Coh(X)) for some smooth and proper variety X, such a result is true only for exact equivalences [46] A convenient localization of DGcat. Waldhausen s K-theory construction K produces a functor DGcat HoSpt, where HoSpt is the (triangulated) homotopy category of spectra. More precisely, given any DG category A one constructs a Waldhausen category structure on the category of perfect right A-modules with cofibrations as module morphisms which admit a (graded) retraction and weak equivalences as quasi-isomorphisms. Then one applies Waldhausen s machinery [65] to this category. The homotopy groups of this spectrum are by definition the Waldausen K-theory groups of the DG category A. We may perform a localization L K (DGcat) with respect to the class of morphisms inverted by K, i.e., morphisms f such that (K(f)) is a (connective) homotopy equivalence of spectra. The category DGcat is combinatorial, i.e., it is cofibrantly generated model category and its underlying category is locally presentable (see, e.g., [1] for the generalities). Indeed, it is cofibrantly generated by construction and the underlying category is locally presentable as follows: if V is a locally presentable symmetric monoidal category then the category of small V-enriched categories is also locally presentable [38] and the category of cochain complexes over k is clearly locally presentable. Now any combinatorial model category is Quillen equivalent to a left proper and simplicial model category [23]. Therefore we may define L K (DGcat) as a Bousfield homological localization [8, 7] and avoid set-theoretic problems. This drastic localization has the effect that all DG categories which are indistinguishable at the level of K-theory spectra up to homotopy become isomorphic in the homotopy category HoL K (DGcat). The category HoL K (DGcat) enjoys the property that the K-theory functors K i : DGcat Ab factor through it, where Ab is the category of abelian groups. 6
8 We call a category semiadditive if it has a zero object, it has finite products and coproducts, the canonical map from the coproduct to the product (which exists thanks to the zero object) is an isomorphism and it is enriched over discrete commutative monoids. Lemma 1.4. The category HoL K (DGcat) is semiadditive. Proof. It was shown in [62] that the homotopy category of DGcat is pointed, i.e, it has a zero object and it has finite products and coproducts. The canonical map from a finite coproduct to the product of DG categories induces an isomorphism between their Waldhausen K-theory spectra, since K is an additive invariant [24, 63]. Therefore, this map is invertible in NCC K dg. The addition (commutative monoid operation) of morphisms f, g : C D is defined by the following composition of arrows C C C (f,g) D D = D D ι D, where is the diagonal map and ι is the fold map, induced by the universal property of the coproduct applied to two copies of the map id A : A A, i.e., it is the dual of the diagonal map. Now we may apply the monoidal group completion functor to the morphism sets (enriched over monoids) of HoL K (DGcat) to obtain certain categories enriched over abelian groups. Since the monoids here are discrete a naïve group completion suffices. Products, coproducts and the zero object remain unaffected. Therefore, by construction we end up with an additive category. We define this additive category as our noncommutative K-correspondence category in this framework and denote it by NCC K dg. There is a canonical functor HoL K (DGcat) NCC K dg which is identity on objects and sends each morphism monoid to its group completion via the canonical map. Since Hmo 0 is the universal additive invariant [63] the functor K on DGcat factors through Hmo 0 and there is a commutative diagram (with additive functors) K Hmo 0 HoSpt K NCC K dg Remark 1.5. An enriched (over HoSpt) version of NCC K dg can be obtained by performing the topological group completion given by ΩB( ), which is the classifying space functor B followed by the loop functor Ω Noncommutative C -K-correspondence category KK C. The category of commutative separable C -algebras corresponds to that of metrizable topological spaces. Kasparov developed KK-theory by unifying K-theory and K-homology into a bivariant theory and obtained interesting positive instances of the Baum Connes conjecture [35, 34]. A remarkable feature of this theory is the existence of an associative Kasparov product on the KK-groups. A categorical point of view of KK-theory making use of the Kasparov product to define compositions was proposed in [28]. The category of C -algebras with morphisms enriched over Kasparov s bivariant KK 0 -groups plays the role of the category of noncommutative 7
9 correspondences in the realm of noncommutative geometry à la Connes (see, for instance, [17, 15]). We denote this category by KK C. Morphisms in the KK 0 -groups can be expressed as homotopy classes of even Kasparov bimodules. Somewhat miraculously in the end one finds that all the analysis disappears and the morphisms are purely determined by topological data. Let us reiterate that it is not quite clear how geometric an abstract C -algebra is. Connes provided a convenient framework of spectral triples to incorporate geometric structures into the picture [14]. A promising candidate for the category of spectral triples has been put forward in [43]. The category KK C may be regarded as a convenient model for the operator algebraic noncommutative K-correspondence category, where most of the well-known geometric examples fit in nicely. There is a canonical functor ι : Sep C KK C, where Sep C is the category of C -algebras with -homomorphisms. The functor ι is identity on objects and makes the target of a -homomorphism a bimodule in the obvious manner. Let K denote the algebra of compact operators on a separable Hilbert space. We set A K := A K. Remark 1.6. There is a counterpart of NCS dg in the world of C -algebras, which we denote by NCS C. For the details we refer the readers to [44], where it was called the category of correspondences in the operator algebraic setting. We regard this category as a category of noncommutative spaces where stably isomorphic algebras are identified. For separable C - algebras being stably isomorphic is equivalent to being Morita Rieffel equivalent [11]. The objects of NCS C are C -algebras and a morphism A B is an isomorphism class of a right Hilbert B K -module E with a nondegenerate -homomorphism f : A K K(E). There is a canonical functor Sep C NCS C which is the universal C -stable functor on Sep C (Proposition 39 loc. cit.). In what follows we shall use KK (resp. K) and KK 0 (resp. K 0 ) interchangeably The passage from KK C to NCC K dg. For any C -algebra A the mapping A A K sending a a π, where π is any rank one projection, is called the corner embedding. This map is clearly nonunital. A functor from Sep C is called C -stable if the image of the corner embedding under the functor is an isomorphism. The definition of an exact sequence in Sep C is simply a diagram isomorphic to 0 I A A/I 0, where I is a closed two-sided ideal in A. Such a diagram is also known as an extension diagram. Since we are working with C -algebras, such extensions are pure, i.e., an inductive limit of k-module split extensions (see Theorem A.4. of [67]). It is further called split exact if it admits a splitting -homomorphism s : A/I A (up to an isomorphism). A functor from Sep C to a Quillen exact category is called split exact if it sends a split exact sequence of C - algebras to a distinguished short exact sequence in the target exact category. An abelian (resp. additive) category admits a natural exact structure, where the distinguished exact sequences are the natural short exact sequences (resp. direct sum diagrams). Higson proved that Kasparov s bivariant K-theory is the universal C -stable and split exact functor from Sep C to an exact category, i.e., given any exact category C and a C -stable and split exact functor F : Sep C C, there is a unique functor F : KK C C such that F ι = F [28, 29]. Such a functor is automatically homotopy invariant [29]. In this section we construct a covariant functor Top K fib : KK C NCCK dg. Our strategy would be to show that the functor Top K fib is a C -stable and a split exact functor on Sep C so that we can apply the universal property of KK-theory to deduce that it factors through the category KK C. 8
10 The category NCS C is the category of C -algebras with some generalized morphisms, in which Morita Rieffel equivalent C -algebras become isomorphic (see Remark 1.6). A C - algebra A is called stable if A = A K, e.g., the algebra of compact operators K is itself stable. In NCS C any C -algebra A is isomorphic to a stable C -algebra functorially, viz., its own stabilization A K. Given a split exact diagram in Sep C 0 A i B j C 0 there exists a morphism t : B A in KK C, which makes it a direct sum diagram. Since KK-theory is morally the space of morphisms between K-theories, we would like our DG category to be a categorical incarnation of K-theory, even for nonunital algebras. Let us briefly recall a construction of Quillen [52], which turns out to be useful to this end. Given any (possibly nonunital) k-algebra A, with unitization Ã, we consider the category Top K dg (A) whose objects are complexes U of right Ã-modules, which are homotopy equivalent to bounded complexes of finitely generated projective modules over Ã, such that U/UA is acyclic. We enrich the category Top K dg (A) over cochain complexes as explained in Example 1.1 to make it a k-linear DG category. So the zeroth cocycle category Z 0 (Top K dg (A)) forms a subcategory of the category of perfect complexes over Ã. Observe that Z0 (Top K dg (A)) is also a Waldhausen category with the weak equivalences (resp. cofibrations) pulled back from the associated Waldhausen category structure on the category of perfect right Top K dg (A)-modules, which are precisely the homotopy equivalences, i.e., maps which become isomorphisms in the homotopy category H 0 (Top K dg (A)), (resp. monomorphisms with a graded splitting). The Grothendieck group of Z 0 (Top K dg (A)) can be identified with the free abelian group generated by the homotopy classes of its objects and relations coming from short exact sequences of complexes, which are split in each degree, i.e., Waldhausen s K 0 -group. In general there is an exact functor from Z 0 (Top K dg (A)) to the Waldhausen category of all perfect complexes over à whose K-theory spectrum is canonically homotopy equivalent to Quillen s algebraic K-theory spectrum (obtained, for instance, by Q-construction). This induces a map of spectra K(Z 0 (Top K dg (A))) Kalg (A). Observe that excision holds for algebraic K-theory of C -algebras [61] and so it makes sense to talk about the algebraic K-theory spectrum of a nonunital C -algebra. As a result there is a canonical map at the level of Grothendieck groups K 0 (Top K dg (A)) K 0(A) := K 0 (Ã)/K 0(k), where K 0 (Ã) (resp. K 0 (k)) can be identified with the Grothendieck group of stable isomorphism classes of finitely generated projective modules over à (resp. k). This map turns out to be an isomorphism when A is a C -algebra (Proposition 6.3 in [52]). For any unital algebra? let hofp(?) denote the category of complexes over it which are homotopy equivalent to a complex of finitely generated projective modules. Then it follows that there is a canonical map from Top K dg (A) to the kernel of the map à Ã/A : hofp(ã) hofp(ã/a) = hofp(k) in DGcat, which becomes an isomorphism in NCC K dg. Any -homomorphism g : A B between possibly nonunital C -algebras extends uniquely to a unital map (preserving the adjoined unit) g : à B between their unitizations. Then one can consider B as an Ã- B-bimodule (left structure is given by the map g), which gives rise to a functor g := à B : Top K dg (A) hofp( B). However, the composition of g 9 s
11 with the map hofp( B) hofp( B/B) = hofp(k) is 0 and hence the image of g lies inside Top K dg (B). In the definition of g we take the algebraic tensor product. Therefore, our construction A Top K dg (A) is functorial with respect to -homomorphisms. Lemma 1.7. The functor Top K dg : Sep C NCCK dg is C -stable. Proof. It is known that if two separable (more generally σ-unital) C -algebras are stably isomorphic then they are Morita Rieffel equivalent [11]. Since A and A K are stably isomorphic there are, by definition, Morita Rieffel equivalence bimodules A X AK and AK X A, satisfying certain conditions, see e.g., [54]. Then X can be made into a left (resp. right) unitary module over the unitization à (resp. à K ) of A (resp. A K ) by setting (a, λ)x = ax + λx for (a, λ) ) Ã. Similarly one makes X into a left ÃK and right Ã-module. One can check that defines a Morita context. Now by Corollary 3.2. of [52] one deduces that X A ( à X X à K and X AK induce inverse equivalences between Top K dg (A) and TopK dg (A K) in NCS dg. In the case when a -homomorphism can be represented as a Morita context, for instance, if we restrict our attention only to -isomorphisms, then using Top K dg we do get isomorphisms in NCS dg via explicit DG functors. Let ( ) gpd denote the underlying groupoid of a category. As a consequence of the above result we obtain a functorial construction at the level of groupoids between noncommutative spaces. Corollary 1.8. There is an induced functor Top K dg : NCS C gpd NCS dg gpd. Remark 1.9. In fact, Theorem 4.2. of [52] asserts that up to a Morita context Top K dg ( ) is independent of the embedding of in a unital C -algebra as a closed two-sided ideal, which, coupled with the main result of [24], ensures that the functor K satisfies excision, i.e., whenever 0 A B C 0 is an exact sequence of C -algebras K(Top K dg (A)) K(Top K dg (B)) K(TopK dg (C)) is a (weak) homotopy fibration. Note that the category NCS dg does not involve any K-theoretic localization. Since the C - stability property of Top K dg is actually achieved in NCS dg gpd it is independent of the K-theoretic localization. The localization will be needed now to prove its split exactness. Lemma The functor Top K dg : Sep C NCCK dg is split exact. Proof. For any split exact sequence applying Top K fib we obtain the diagram 0 A i B j C 0, s S=s! Top K dg (A) I=i! Top K dg (B) J=j! Top K dg (C), where JI = 0 and JS = id Top K dg (C). One can construct ker(j) in the Karoubian closure of NCC K dg, because SJ : Top K dg (B) TopK dg (B) is a projection, and obtain the following diagram 10
12 (1) Top K dg (A) I S κ 0 I ker(j) Top K dg (B) J Top K dg (C) where the existence of κ follows from the universal properties of ker(j). Our aim is to show that κ is an isomorphism in NCC K dg. We may apply Waldhausen s K-theory spectrum functor K to the above diagram (1) and by Remark 1.9 it follows that K(κ) is an isomorphism in the homotopy category of spectra, whence κ is an isomorphism in NCC K dg. Now for some technical benefits we modify our functor Top K dg slightly. For any C -algebra we define Top K fib ( ) = per dg(top K dg ( )) (see Definition 1.2 and the remark thereafter). The functor per dg is simply the fibrant replacement functor in DGcat, the canonical Yoneda map θ : per dg ( ) being an isomorphism. Hence it induces a homotopy equivalence at the level of K-theory spectra. A map f : in Sep C induces an unnatural map f! : per dg (Top K dg ( )) per dg(top K dg ( )) (in the same direction) by setting f! = θ f θ 1. Therefore, Top K fib ( ) is a covariant functor Sep C NCCK dg. Remark For any unital C -algebra the category Top K fib ( ) is our DG model (up to a derived Morita equivalence) for the bounded derived category of finitely generated projective right -modules. Remark Observe that by definition the image of Top K fib is a k-linear pretriangulated DG category. The advantage of pretriangulated DG categories is that one can construct cones and cylinders of morphisms functorially in them. The application of per dg also creates some flexibility to bring in more analysis into the picture. For instance, instead of taking DG functors with values in chain complexes over k one could take functors with values in the category of chain complexes over the quasiabelian category of topological vector spaces [58, 51]. Conceivably one could still prove a result similar to Theorem 1.16 below, which we leave for the readers to figure out. Lemma The functor Top K fib : Sep C NCCK dg is C -stable and split exact. Proof. The assertions follow from Lemma 1.7 and Lemma 1.10 since per dg ( ) is simply a fibrant replacement of in the Morita model category DGcat. Lemma The functor Top K fib : Sep C NCCK dg is homotopy invariant. Proof. This is an immediate consequence of Theorem of [29] which says that any C -stable and split exact functor on Sep C is automatically homotopy invariant. Now we prove a Proposition which shows that our functor Top K fib encodes topological K- theory. The proof is modelled along the lines of ibid.. Proposition Let A be any C -algebra. Then K(Top K fib (A)) is homotopy equivalent to the connective cover K top (A) 0 of the topological K-theory spectrum. 11 0,
13 Proof. By Lemma 1.13 we may replace A by A K and use the fact that Top K dg ( ) = Top K fib ( ) in NCS dg. For the benefit of the reader we now recall a standard dimension shifting argument for stable C -algebras using the exact sequence 0 C 0 ((0, 1)) A K C 0 ([0, 1)) A K A K 0. As discussed above there is a map of spectra K(Top K dg (A K)) K alg (A K ) which induces an isomorphism at the level of K 0. Using Remark 1.9 we obtain the following map of exact sequences [set K dg i ( ) = π i (K(Top K dg ( ))), C 0([0, 1)) A K ) = ConeA K and C 0 ((0, 1)) = Σ ] K dg 1 (ConeA K ) K dg 1 (A K ) K dg 0 (ΣA K ) K dg 0 (ConeA K ) K dg 0 (A K ) K alg 1 (ConeA K ) K alg 1 (A K ) K alg 0 (ΣA K ) K alg 0 (ConeA K ) K alg 0 (A K ). We know that the three vertical arrows from the right are isomorphisms. Now we exploit the homotopy invariance of K dg i and K alg i and the fact that ConeA K is contractible (see, e.g., Theorem of [29]) to deduce that K dg i (ConeA K ) = K alg i (ConeA K ) = 0, whence the boundary maps K alg 1 (A K ) K alg 0 (ΣA K ) and K dg 1 (A K ) K dg 0 (ΣA K ) are isomorphisms. It follows immediately that K dg 1 (A K ) K alg 1 (A K ) is an isomorphism. The isomorphisms K dg i (A K ) K alg i (A K ) for i 2 follow easily by induction. Thanks to the Theorem of Suslin Wodzicki [61] we know that the algebraic K-theory spectrum is (connectively) homotopy equivalent to the topological K-theory spectrum of a stable C -algebra (see also, e.g., Theorem 1.4. of [55]). Since A K is stable, K(Top K dg (A K)) is actually homotopy equivalent to the connective cover of the topological K-theory spectrum of A K, which in turn is homotopy equivalent to K top (A) 0. Now we state the main Theorem in this section. Theorem The functor Top K fib factors through KK C ; in other words, we have the following commutative diagram of functors: Top K fib Sep C NCC K dg. ι Top K fib KK C Proof. We have already checked that the functor Top K fib is C -stable (Lemma 1.13) and split exact (Lemma 1.10). It remains to apply Higson s characterization of KK as the universal C -stable and split exact functor on Sep C [28, 29]. Corollary An isomorphism in KK C implies an isomorphism in NCC K dg. In other words, KK-equivalence implies (K-correspondence like) derived DG Morita equivalence up to a K- theoretic identification, or Morita-K equivalence, for brevity. It is known that two unital rings with equivalent derived categories of modules have isomorphic (algebraic) K-theories [24]. We have the following sequence of implications for C -algebras Morita Rieffel equiv. KK-equiv. Morita-K equiv. isom. top. K-theories. 12
14 Remark Any C -stable and split exact functor on Sep C satisfies Bott periodicity [29] (see also [21]). Hence the functor Top K fib will also have this property. It is useful to know that the functor Top K fib : KK C NCCK dg exists by abstract reasoning. However, in order to make the situation a bit more transparent we make use of a rather algebraic formulation of KK-theory [19]. For any C -algebra A let A A denote the free product (which is the coproduct in Sep C ) of two copies of A and let qa be the kernel of the fold map A A A. It was shown in ibid. that A (resp. B) is isomorphic to qa (resp. B K ) in KK C and KK(A, B) = [qa, B K ], i.e., homotopy classes of -homomorphisms qa B K. Roughly, the algebra qa is expected to play the role of a cofibrant replacement of A and B K that of a fibrant replacement of B with respect to some model structure with KK - equivalences as weak equivalences. This goal has been accomplished in a larger category [33], where all objects are fibrant and a minor modification of qa acts as a cofibrant replacement [33]. One benefit of this approach is that Kasparov s product can be viewed simply as a composition of -homomorphisms, which is quite often easier to deal with. In order to define the abelian group structure one proceeds roughly as follows: for any φ, ψ [qa, B K ] one defines φ ψ : qa M 2 (B K ) as ( ) φ 0 0 ψ. Then one argues that M2 (B K ) is isomorphic to B K in KK C and fixing such an isomorphism θ one sets φ + ψ := θ(φ ψ). We can exploit this fact by concluding that if A is isomorphic to B in KK C then Top K fib (qa) is isomorphic to Top K fib (B K) in NCC K dg by an explicit DG functor induced by the -homomorphism qa B K that is invertible in KK C. Proposition The functor Top K fib : KK C NCCK dg is additive. Proof. We simply use the fact that any direct sum diagram in KK C can be expressed as a split exact diagram involving only -homomorphisms applying q( ) and K several times, both of which produce isomorphic objects in KK C, and then apply the split exactness property of Top K fib. Example Let A = C(E) be the C -algebra of continuous functions on a complex elliptic curve E. Topologically E is isomorphic to a 2-torus T 2. It is known that K(A) is isomorphic to Z 2. Using the fact that A belongs to the Universal Coefficient Theorem class one computes KK(A, A) M(2, Z). The group of invertible elements can be identified with GL(2, Z). Let D b (E) be the bounded derived category of coherent sheaves on E. Since all autoequivalences of D b (E) are geometric in nature (see, e.g., Theorem of [46]), they definitely give rise to automorphisms of Top K fib (A); in other words, there is a group homomorphism Aut(Db (E)) Aut(Top K fib (A)). The automorphism group Aut(D b (E)) can be described explicitly (see, e.g., Remark (iv) [12]). It maps surjectively onto SL(2, Z) with a non-canonical splitting defined by sending the generators of SL(2, Z) to some specific Seidel Thomas twist functors. It is clear that the group Aut(D b (E)) is bigger than GL(2, Z) since it contains Pic 0 (E) as a subgroup. The SL(2, Z) part of Aut(D b (E)) can be described by the Seidel Thomas twist functors, which can also be seen as Fourier Mukai transforms [59] and it seems that KK-equivalences can account for them. The automorphism groups of commutative C -algebras in KK C are computable from their K-theories using the Universal Coefficient Theorem [57]. As the example suggests, for any 13
15 locally compact Hausdorff topological space X the group Aut KKC (C 0 (X)) always maps to Aut NCC K dg (Top K fib (C 0(X)) and gives some idea about the automorphisms of the DG derived category of topological vector bundles of X. However, the automorphism group in NCC K dg will typically be larger than the automorphism group of derived categories consisting of exact equivalences up to a natural transformation E-theory and strong deformations up to homotopy. The Connes Higson E- theory [16] is the universal C -stable, exact and homotopy invariant functor. Recall that a functor F is exact if it sends an exact sequence of C -algebras 0 A B C 0 to an exact sequence F (A) F (B) F (C) (exact at F (B)). It is known that any exact and homotopy invariant functor is split-exact. Therefore, there is a canonical induced functor KK C E, where the category E consists of C -algebras with bivariant E 0 -groups as morphisms. This functor is fully faithful when restricted to nuclear C -algebras; in fact, by the Choi Effros lifting Theorem KK (A, B) = E (A, B) whenever A is nuclear. Thus, restricted to nuclear C -algebras there are maps E 0 (A, B) = KK 0 (A, B) Hom NCC K dg (Top K fib (A), TopK fib (B)). Let B := C b ([1, ), B)/C 0 ([1, ), B), where C b denotes bounded continuous functions, be the asymptotic algebra of B. An asymptotic morphism between C -algebras A and B is a -homomorphisms φ : A B. A strong deformation of C -algebras from A to B is a continuous field A(t) of C -algebras over [0, 1] whose fibre at 0, A(0) = A and whose restriction to (0, 1] is the constant field with fibre A(t) = B for all t (0, 1]. Given any such strong deformation of C -algebras and an a A one can choose a section α a (t) of the continuous field such that α a (0) = a. Suppose one has chosen such a section α a (t) for every a A. Then one associates an asymptotic morphism by setting (φ(a))(t) = α a (1/t), t [1, ). Let ΣA := C 0 ((0, 1), A) be the suspension of A. Then the Connes Higson picture of E-theory says that E 0 (A, B) = [[ΣA K, ΣB K ]], where [[?, ]] denotes homotopy classes of asymptotic morphisms between? and. Let φ be an asymptotic morphism defined by a strong deformation from A to B. Then we call the class of φ in E 0 (A, B) a strong deformation up to homotopy from A to B. If A is nuclear, e.g., if A is commutative, whenever the class of φ is invertible in E 0 (A, B), we deduce that Top K fib (A) and Top K fib (B) are isomorphic in NCCK dg Homological T-dualities. A sigma model roughly studies maps Σ X, where Σ is called the worldsheet (Riemann surface) and X the target spacetime (typically a 10- dimensional manifold in supersymmetric string theories). Mirror symmetry relates the sigma models of type IIA and IIB string theories with dual Calabi Yau target spacetimes. In open string theories, i.e., when Σ has boundaries, the boundaries are constrained to live in some special submanifolds of the spacetime X. Such a submanifold also comes equipped with a special Chan-Paton vector bundle and together they define a topological K-theory class of X via the Gysin map. The D-brane charges correspond to such K-theory classes [66]. The homological mirror symmetry conjecture of Kontsevich predicts an equivalence of triangulated categories of IIA-branes (Fukaya category) on a Calabi Yau target manifold X and IIBbranes (derived category of coherent sheaves) on its dual ˆX. This equivalence would induce an isomorphism between their Grothendieck groups and it was argued that the Grothedieck group of the category of A-branes on X, at least when the dimension of X is not divisible by 4, should be isomorphic to K 1 top(x) [31]. Strominger Yau Zaslow argued that sometimes 14
16 when X and ˆX are mirror dual Calabi Yau 3-folds one should be able to find a generically T 3 -fibration over a common base Z (2) X T-duality ˆX, Z such that mirror symmetry is obtained by applying T-duality fibrewise [60]. Since T- duality is applied an odd number of times it interchanges types (IIA IIB). Sometimes using Poincaré duality type arguments it is possible to identify topological K-theory with K-homology. Kasparov s KK-theory naturally subsumes K-theory and K-homology and it was shown in [10] that certain topological T-duality transformations (even including more parameters like H-fluxes, which we did not discuss here) can naturally be seen as KK 1 - classes between suitably defined continuous trace C -algebras capturing the geometry of the above diagram 2. As argued above, whilst an odd number of T-duality transformations interchanges types (IIA IIB), an even number preserves it and corresponds to a KK 0 -class. Therefore, a topological C -K-correspondences or a KK 0 -class is an abstract generalization of an even number of T-duality transformations (or T 2n -dualities), viewed as an equivalence of IIB-branes (or IIA-branes) on the same target manifold and inducing an isomorphism at the level of K-theory. Since KK-theory is Bott periodic one can also use the identification KK 1 (, ) = KK 0 (, C 0 ((0, 1)) ). 2. Simplicial sets and pro C -algebras In this section we construct a pro C -algebra from a simplicial set and show that the construction is functorial with respect to proper maps between simplicial sets and pro C - algebras. We also show that this construction respects homotopy of proper maps after stabilizing the category of pro C -algebras with respect to finite matrices Generalities on simplicial sets and pro C -algebras. The standard reference for simplicial aspects of topology is [27]. Let be the cosimplicial category, i.e., the category whose objects are finite ordinals [n] := {0, 1,, n} and whose morphisms are monotonic nondecreasing maps. Its morphisms admit a unique decomposition in terms of coface and codegeneracy maps, which satisfy certain well-known relations. Let Set be the category of all sets. By a simplicial set we mean a functor Σ : op Set and a morphism of simplicial sets is a natural transformation between these functors. We denote by SSet the category of simplicial sets. The elements of Σ[n] are called the n-simplices (or n-dimensional simplices) and the images of the coface and codegeneracy maps in are called the face (denoted by d i ) and degeneracy maps (denoted by s j ). An n-simplex σ is called degenerate if it is of the form σ = s i (τ) for some (n 1)-simplex τ. Degenerate simplices are needed to ensure that maps of graded sets exist, even if the target simplicial set has no nondegenerate simplex in a particular dimension. Simplicial sets provide a combinatorial description of topology. The singular simplices functor and the geometric realization functor are adjoint functors between the category of compactly generated and Hausdorff topological spaces and that of simplicial sets, which induce inverse equivalences between their homotopy categories with respect to their natural model category structures. 15
An introduction to derived and triangulated categories. Jon Woolf
An introduction to derived and triangulated categories Jon Woolf PSSL, Glasgow, 6 7th May 2006 Abelian categories and complexes Derived categories and functors arise because 1. we want to work with complexes
More informationBernhard Keller. University Paris 7 and Jussieu Mathematics Institute. On differential graded categories. Bernhard Keller
graded graded University Paris 7 and Jussieu Mathematics Institute graded Philosophy graded Question: What is a non commutative (=NC) scheme? Grothendieck, Manin,... : NC scheme = abelian category classical
More informationNoncommutative geometry and quantum field theory
Noncommutative geometry and quantum field theory Graeme Segal The beginning of noncommutative geometry is the observation that there is a rough equivalence contravariant between the category of topological
More informationarxiv: v1 [math.at] 17 Apr 2008
DG CATEGORIES AS EILENBERG-MAC LANE SPECTRAL ALGEBRA arxiv:0804.2791v1 [math.at] 17 Apr 2008 GONÇALO TABUADA Abstract. We construct a zig-zag of Quillen equivalences between the homotopy theories of differential
More informationIndCoh Seminar: Ind-coherent sheaves I
IndCoh Seminar: Ind-coherent sheaves I Justin Campbell March 11, 2016 1 Finiteness conditions 1.1 Fix a cocomplete category C (as usual category means -category ). This section contains a discussion of
More informationRTG Mini-Course Perspectives in Geometry Series
RTG Mini-Course Perspectives in Geometry Series Jacob Lurie Lecture IV: Applications and Examples (1/29/2009) Let Σ be a Riemann surface of genus g, then we can consider BDiff(Σ), the classifying space
More informationINTRODUCTION TO PART V: CATEGORIES OF CORRESPONDENCES
INTRODUCTION TO PART V: CATEGORIES OF CORRESPONDENCES 1. Why correspondences? This part introduces one of the two main innovations in this book the (, 2)-category of correspondences as a way to encode
More informationSymplectic geometry of homological algebra
Symplectic geometry of homological algebra Maxim Kontsevich June 10, 2009 Derived non-commutative algebraic geometry With any scheme X over ground field k we can associate a k-linear triangulated category
More informationThree Descriptions of the Cohomology of Bun G (X) (Lecture 4)
Three Descriptions of the Cohomology of Bun G (X) (Lecture 4) February 5, 2014 Let k be an algebraically closed field, let X be a algebraic curve over k (always assumed to be smooth and complete), and
More information1 Notations and Statement of the Main Results
An introduction to algebraic fundamental groups 1 Notations and Statement of the Main Results Throughout the talk, all schemes are locally Noetherian. All maps are of locally finite type. There two main
More informationHomological mirror symmetry via families of Lagrangians
Homological mirror symmetry via families of Lagrangians String-Math 2018 Mohammed Abouzaid Columbia University June 17, 2018 Mirror symmetry Three facets of mirror symmetry: 1 Enumerative: GW invariants
More informationSymmetric monoidal structure on Non-commutative motives
Symmetric monoidal structure on Non-commutative motives Denis-Charles Cisinski, Gonçalo Tabuada To cite this version: Denis-Charles Cisinski, Gonçalo Tabuada. Symmetric monoidal structure on Non-commutative
More informationarxiv: v2 [math.at] 18 Sep 2008
TOPOLOGICAL HOCHSCHILD AND CYCLIC HOMOLOGY FOR DIFFERENTIAL GRADED CATEGORIES arxiv:0804.2791v2 [math.at] 18 Sep 2008 GONÇALO TABUADA Abstract. We define a topological Hochschild (THH) and cyclic (TC)
More informationMORITA HOMOTOPY THEORY OF C -CATEGORIES IVO DELL AMBROGIO AND GONÇALO TABUADA
MORITA HOMOTOPY THEORY OF C -CATEGORIES IVO DELL AMBROGIO AND GONÇALO TABUADA Abstract. In this article we establish the foundations of the Morita homotopy theory of C -categories. Concretely, we construct
More informationDe Rham Cohomology. Smooth singular cochains. (Hatcher, 2.1)
II. De Rham Cohomology There is an obvious similarity between the condition d o q 1 d q = 0 for the differentials in a singular chain complex and the condition d[q] o d[q 1] = 0 which is satisfied by the
More informationAXIOMS FOR GENERALIZED FARRELL-TATE COHOMOLOGY
AXIOMS FOR GENERALIZED FARRELL-TATE COHOMOLOGY JOHN R. KLEIN Abstract. In [Kl] we defined a variant of Farrell-Tate cohomology for a topological group G and any naive G-spectrum E by taking the homotopy
More informationOperator Algebras II, Homological functors, derived functors, Adams spectral sequence, assembly map and BC for compact quantum groups.
II, functors, derived functors, Adams spectral, assembly map and BC for compact quantum groups. University of Copenhagen 25th July 2016 Let C be an abelian category We call a covariant functor F : T C
More informationAlgebraic Geometry Spring 2009
MIT OpenCourseWare http://ocw.mit.edu 18.726 Algebraic Geometry Spring 2009 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. 18.726: Algebraic Geometry
More informationStrongly Self-Absorbing C -algebras which contain a nontrivial projection
Münster J. of Math. 1 (2008), 99999 99999 Münster Journal of Mathematics c Münster J. of Math. 2008 Strongly Self-Absorbing C -algebras which contain a nontrivial projection Marius Dadarlat and Mikael
More informationIntroduction to the Baum-Connes conjecture
Introduction to the Baum-Connes conjecture Nigel Higson, John Roe PSU NCGOA07 Nigel Higson, John Roe (PSU) Introduction to the Baum-Connes conjecture NCGOA07 1 / 15 History of the BC conjecture Lecture
More informationSECTION 5: EILENBERG ZILBER EQUIVALENCES AND THE KÜNNETH THEOREMS
SECTION 5: EILENBERG ZILBER EQUIVALENCES AND THE KÜNNETH THEOREMS In this section we will prove the Künneth theorem which in principle allows us to calculate the (co)homology of product spaces as soon
More information6. Lecture cdh and Nisnevich topologies. These are Grothendieck topologies which play an important role in Suslin-Voevodsky s approach to not
6. Lecture 6 6.1. cdh and Nisnevich topologies. These are Grothendieck topologies which play an important role in Suslin-Voevodsky s approach to not only motivic cohomology, but also to Morel-Voevodsky
More informationp,q H (X), H (Y ) ), where the index p has the same meaning as the
There are two Eilenberg-Moore spectral sequences that we shall consider, one for homology and the other for cohomology. In contrast with the situation for the Serre spectral sequence, for the Eilenberg-Moore
More informationUniversal Coefficient Theorems and assembly maps in KK-theory
Universal Coefficient Theorems and assembly maps in KK-theory Ralf Meyer Abstract. We introduce equivariant Kasparov theory using its universal property and construct the Baum Connes assembly map by localising
More informationThus we get. ρj. Nρj i = δ D(i),j.
1.51. The distinguished invertible object. Let C be a finite tensor category with classes of simple objects labeled by a set I. Since duals to projective objects are projective, we can define a map D :
More informationA GLIMPSE OF ALGEBRAIC K-THEORY: Eric M. Friedlander
A GLIMPSE OF ALGEBRAIC K-THEORY: Eric M. Friedlander During the first three days of September, 1997, I had the privilege of giving a series of five lectures at the beginning of the School on Algebraic
More informationGrothendieck duality for affine M 0 -schemes.
Grothendieck duality for affine M 0 -schemes. A. Salch March 2011 Outline Classical Grothendieck duality. M 0 -schemes. Derived categories without an abelian category of modules. Computing Lf and Rf and
More informationProper Actions on C -algebras
Dartmouth College CBMS TCU May 2009 This is joint work with Astrid an Huef, Steve Kaliszewski and Iain Raeburn Proper Actions on Spaces Definition We say that G acts properly on a locally compact space
More informationT-duality & noncommutative geometry
T-duality & noncommutative geometry Type IIA Type IIB duality rephrased Higher Structures in String Theory and Quantum Field Theory Instructional workshop for students and junior researchers Mathai Varghese
More informationPART II.1. IND-COHERENT SHEAVES ON SCHEMES
PART II.1. IND-COHERENT SHEAVES ON SCHEMES Contents Introduction 1 1. Ind-coherent sheaves on a scheme 2 1.1. Definition of the category 2 1.2. t-structure 3 2. The direct image functor 4 2.1. Direct image
More informationTHE CELLULARIZATION PRINCIPLE FOR QUILLEN ADJUNCTIONS
THE CELLULARIZATION PRINCIPLE FOR QUILLEN ADJUNCTIONS J. P. C. GREENLEES AND B. SHIPLEY Abstract. The Cellularization Principle states that under rather weak conditions, a Quillen adjunction of stable
More informationThe Kervaire Invariant One Problem, Talk 0 (Introduction) Independent University of Moscow, Fall semester 2016
The Kervaire Invariant One Problem, Talk 0 (Introduction) Independent University of Moscow, Fall semester 2016 January 3, 2017 This is an introductory lecture which should (very roughly) explain what we
More informationEXPANDERS, EXACT CROSSED PRODUCTS, AND THE BAUM-CONNES CONJECTURE. 1. Introduction
EXPANDERS, EXACT CROSSED PRODUCTS, AND THE BAUM-CONNES CONJECTURE PAUL BAUM, ERIK GUENTNER, AND RUFUS WILLETT Abstract. We reformulate the Baum-Connes conjecture with coefficients by introducing a new
More informationON THE HOMOTOPY THEORY OF ENRICHED CATEGORIES
ON THE HOMOTOPY THEORY OF ENRICHED CATEGORIES CLEMENS BERGER AND IEKE MOERDIJK Abstract. We give sufficient conditions for the existence of a Quillen model structure on small categories enriched in a given
More informationCrossed products of irrational rotation algebras by finite subgroups of SL 2 (Z) are AF
Crossed products of irrational rotation algebras by finite subgroups of SL 2 (Z) are AF N. Christopher Phillips 7 May 2008 N. Christopher Phillips () C (Z k, A θ ) is AF 7 May 2008 1 / 36 The Sixth Annual
More informationBERTRAND GUILLOU. s G q+r
STABLE A 1 -HOMOTOPY THEORY BERTRAND GUILLOU 1. Introduction Recall from the previous talk that we have our category pointed A 1 -homotopy category Ho A 1, (k) over a field k. We will often refer to an
More informationarxiv: v1 [math.kt] 18 Dec 2009
EXCISION IN HOCHSCHILD AND CYCLIC HOMOLOGY WITHOUT CONTINUOUS LINEAR SECTIONS arxiv:0912.3729v1 [math.kt] 18 Dec 2009 RALF MEYER Abstract. We prove that continuous Hochschild and cyclic homology satisfy
More informationThe Riemann-Roch Theorem
The Riemann-Roch Theorem TIFR Mumbai, India Paul Baum Penn State 7 August, 2015 Five lectures: 1. Dirac operator 2. Atiyah-Singer revisited 3. What is K-homology? 4. Beyond ellipticity 5. The Riemann-Roch
More informationThe Riemann-Roch Theorem
The Riemann-Roch Theorem Paul Baum Penn State Texas A&M University College Station, Texas, USA April 4, 2014 Minicourse of five lectures: 1. Dirac operator 2. Atiyah-Singer revisited 3. What is K-homology?
More informationHomological mirror symmetry
Homological mirror symmetry HMS (Kontsevich 1994, Hori-Vafa 2000, Kapustin-Li 2002, Katzarkov 2002,... ) relates symplectic and algebraic geometry via their categorical structures. A symplectic manifold
More informationTopological K-theory, Lecture 3
Topological K-theory, Lecture 3 Matan Prasma March 2, 2015 1 Applications of the classification theorem continued Let us see how the classification theorem can further be used. Example 1. The bundle γ
More informationHomological Algebra and Differential Linear Logic
Homological Algebra and Differential Linear Logic Richard Blute University of Ottawa Ongoing discussions with Robin Cockett, Geoff Cruttwell, Keith O Neill, Christine Tasson, Trevor Wares February 24,
More informationThe Gauss-Manin Connection for the Cyclic Homology of Smooth Deformations, and Noncommutative Tori
The Gauss-Manin Connection for the Cyclic Homology of Smooth Deformations, and Noncommutative Tori Allan Yashinski Abstract Given a smooth deformation of topological algebras, we define Getzler s Gauss-Manin
More informationPART III.3. IND-COHERENT SHEAVES ON IND-INF-SCHEMES
PART III.3. IND-COHERENT SHEAVES ON IND-INF-SCHEMES Contents Introduction 1 1. Ind-coherent sheaves on ind-schemes 2 1.1. Basic properties 2 1.2. t-structure 3 1.3. Recovering IndCoh from ind-proper maps
More informationo ALGEBRAIC K-THEORY AND HIGHER CATEGORIES ANDREW J. BLUMBERG Abstract. The outline of the talk. 1. Setup Goal: Explain algebraic K-theory as a functor from the homotopical category of homotopical categories
More informationNoncommutative deformations and perverse schobers
Noncommutative deformations and perverse schobers Work in progress with Ludmil Katzarkov Andrew Harder University of Miami January 26, 2017 Andrew Harder (University of Miami) NCD and PSC January 26, 2017
More informationGK-SEMINAR SS2015: SHEAF COHOMOLOGY
GK-SEMINAR SS2015: SHEAF COHOMOLOGY FLORIAN BECK, JENS EBERHARDT, NATALIE PETERNELL Contents 1. Introduction 1 2. Talks 1 2.1. Introduction: Jordan curve theorem 1 2.2. Derived categories 2 2.3. Derived
More informationAlgebraic Cobordism Lecture 1: Complex cobordism and algebraic cobordism
Algebraic Cobordism Lecture 1: Complex cobordism and algebraic cobordism UWO January 25, 2005 Marc Levine Prelude: From homotopy theory to A 1 -homotopy theory A basic object in homotopy theory is a generalized
More informationWe then have an analogous theorem. Theorem 1.2.
1. K-Theory of Topological Stacks, Ryan Grady, Notre Dame Throughout, G is sufficiently nice: simple, maybe π 1 is free, or perhaps it s even simply connected. Anyway, there are some assumptions lurking.
More information121B: ALGEBRAIC TOPOLOGY. Contents. 6. Poincaré Duality
121B: ALGEBRAIC TOPOLOGY Contents 6. Poincaré Duality 1 6.1. Manifolds 2 6.2. Orientation 3 6.3. Orientation sheaf 9 6.4. Cap product 11 6.5. Proof for good coverings 15 6.6. Direct limit 18 6.7. Proof
More informationDerived Algebraic Geometry III: Commutative Algebra
Derived Algebraic Geometry III: Commutative Algebra May 1, 2009 Contents 1 -Operads 4 1.1 Basic Definitions........................................... 5 1.2 Fibrations of -Operads.......................................
More informationCategory Theory. Categories. Definition.
Category Theory Category theory is a general mathematical theory of structures, systems of structures and relationships between systems of structures. It provides a unifying and economic mathematical modeling
More informationin path component sheaves, and the diagrams
Cocycle categories Cocycles J.F. Jardine I will be using the injective model structure on the category s Pre(C) of simplicial presheaves on a small Grothendieck site C. You can think in terms of simplicial
More informationCATEGORICAL GROTHENDIECK RINGS AND PICARD GROUPS. Contents. 1. The ring K(R) and the group Pic(R)
CATEGORICAL GROTHENDIECK RINGS AND PICARD GROUPS J. P. MAY Contents 1. The ring K(R) and the group Pic(R) 1 2. Symmetric monoidal categories, K(C), and Pic(C) 2 3. The unit endomorphism ring R(C ) 5 4.
More informationGroupoids and Orbifold Cohomology, Part 2
Groupoids and Orbifold Cohomology, Part 2 Dorette Pronk (with Laura Scull) Dalhousie University (and Fort Lewis College) Groupoidfest 2011, University of Nevada Reno, January 22, 2012 Motivation Orbifolds:
More informationIterated Bar Complexes of E-infinity Algebras and Homology Theories
Iterated Bar Complexes of E-infinity Algebras and Homology Theories BENOIT FRESSE We proved in a previous article that the bar complex of an E -algebra inherits a natural E -algebra structure. As a consequence,
More informationNoncommutative motives and their applications
MSRI 2013 The classical theory of pure motives (Grothendieck) V k category of smooth projective varieties over a field k; morphisms of varieties (Pure) Motives over k: linearization and idempotent completion
More information1 Motivation. If X is a topological space and x X a point, then the fundamental group is defined as. the set of (pointed) morphisms from the circle
References are: [Szamuely] Galois Groups and Fundamental Groups [SGA1] Grothendieck, et al. Revêtements étales et groupe fondamental [Stacks project] The Stacks Project, https://stacks.math.columbia. edu/
More informationMath 210B. Profinite group cohomology
Math 210B. Profinite group cohomology 1. Motivation Let {Γ i } be an inverse system of finite groups with surjective transition maps, and define Γ = Γ i equipped with its inverse it topology (i.e., the
More informationDerived Algebraic Geometry I: Stable -Categories
Derived Algebraic Geometry I: Stable -Categories October 8, 2009 Contents 1 Introduction 2 2 Stable -Categories 3 3 The Homotopy Category of a Stable -Category 6 4 Properties of Stable -Categories 12 5
More informationAlgebraic models for higher categories
Algebraic models for higher categories Thomas Nikolaus Fachbereich Mathematik, Universität Hamburg Schwerpunkt Algebra und Zahlentheorie Bundesstraße 55, D 20 146 Hamburg Abstract We introduce the notion
More informationHomology and Cohomology of Stacks (Lecture 7)
Homology and Cohomology of Stacks (Lecture 7) February 19, 2014 In this course, we will need to discuss the l-adic homology and cohomology of algebro-geometric objects of a more general nature than algebraic
More informationCW-complexes. Stephen A. Mitchell. November 1997
CW-complexes Stephen A. Mitchell November 1997 A CW-complex is first of all a Hausdorff space X equipped with a collection of characteristic maps φ n α : D n X. Here n ranges over the nonnegative integers,
More information1 Categorical Background
1 Categorical Background 1.1 Categories and Functors Definition 1.1.1 A category C is given by a class of objects, often denoted by ob C, and for any two objects A, B of C a proper set of morphisms C(A,
More informationCGP DERIVED SEMINAR GABRIEL C. DRUMMOND-COLE
CGP DERIVED SEMINAR GABRIEL C. DRUMMOND-COLE 1. January 16, 2018: Byunghee An, bar and cobar Today I am going to talk about bar and cobar constructions again, between categories of algebras and coalgebras.
More informationEILENBERG-ZILBER VIA ACYCLIC MODELS, AND PRODUCTS IN HOMOLOGY AND COHOMOLOGY
EILENBERG-ZILBER VIA ACYCLIC MODELS, AND PRODUCTS IN HOMOLOGY AND COHOMOLOGY CHRIS KOTTKE 1. The Eilenberg-Zilber Theorem 1.1. Tensor products of chain complexes. Let C and D be chain complexes. We define
More informationHochschild homology and Grothendieck Duality
Hochschild homology and Grothendieck Duality Leovigildo Alonso Tarrío Universidade de Santiago de Compostela Purdue University July, 1, 2009 Leo Alonso (USC.es) Hochschild theory and Grothendieck Duality
More informationTopics in Geometry: Mirror Symmetry
MIT OpenCourseWare http://ocw.mit.edu 18.969 Topics in Geometry: Mirror Symmetry Spring 2009 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. MIRROR SYMMETRY:
More informationPART II.2. THE!-PULLBACK AND BASE CHANGE
PART II.2. THE!-PULLBACK AND BASE CHANGE Contents Introduction 1 1. Factorizations of morphisms of DG schemes 2 1.1. Colimits of closed embeddings 2 1.2. The closure 4 1.3. Transitivity of closure 5 2.
More informationSECTION 2: THE COMPACT-OPEN TOPOLOGY AND LOOP SPACES
SECTION 2: THE COMPACT-OPEN TOPOLOGY AND LOOP SPACES In this section we will give the important constructions of loop spaces and reduced suspensions associated to pointed spaces. For this purpose there
More informationON A LOCALIZATION FORMULA OF EPSILON FACTORS VIA MICROLOCAL GEOMETRY
ON A LOCALIZATION FORMULA OF EPSILON FACTORS VIA MICROLOCAL GEOMETRY TOMOYUKI ABE AND DEEPAM PATEL Abstract. Given a lisse l-adic sheaf G on a smooth proper variety X and a lisse sheaf F on an open dense
More informationCATEGORICAL ASPECTS OF ALGEBRAIC GEOMETRY IN MIRROR SYMMETRY ABSTRACTS
CATEGORICAL ASPECTS OF ALGEBRAIC GEOMETRY IN MIRROR SYMMETRY Alexei Bondal (Steklov/RIMS) Derived categories of complex-analytic manifolds Alexender Kuznetsov (Steklov) Categorical resolutions of singularities
More information(G; A B) βg Tor ( K top
GOING-DOWN FUNCTORS, THE KÜNNETH FORMULA, AND THE BAUM-CONNES CONJECTURE. JÉRÔME CHABERT, SIEGFRIED ECHTERHOFF, AND HERVÉ OYONO-OYONO Abstract. We study the connection between the Baum-Connes conjecture
More informationDeformation theory of representable morphisms of algebraic stacks
Deformation theory of representable morphisms of algebraic stacks Martin C. Olsson School of Mathematics, Institute for Advanced Study, 1 Einstein Drive, Princeton, NJ 08540, molsson@math.ias.edu Received:
More informationMatilde Marcolli and Goncalo Tabuada. Noncommutative numerical motives and the Tannakian formalism
Noncommutative numerical motives and the Tannakian formalism 2011 Motives and Noncommutative motives Motives (pure): smooth projective algebraic varieties X cohomology theories H dr, H Betti, H etale,...
More informationLOOP SPACES IN MOTIVIC HOMOTOPY THEORY. A Dissertation MARVIN GLEN DECKER
LOOP SPACES IN MOTIVIC HOMOTOPY THEORY A Dissertation by MARVIN GLEN DECKER Submitted to the Office of Graduate Studies of Texas A&M University in partial fulfillment of the requirements for the degree
More informationsset(x, Y ) n = sset(x [n], Y ).
1. Symmetric monoidal categories and enriched categories In practice, categories come in nature with more structure than just sets of morphisms. This extra structure is central to all of category theory,
More informationA Taxonomy of 2d TQFTs
1 Section 2 Ordinary TFTs and Extended TFTs A Taxonomy of 2d TQFTs Chris Elliott October 28, 2013 1 Introduction My goal in this talk is to explain several extensions of ordinary TQFTs in two dimensions
More informationA TALE OF TWO FUNCTORS. Marc Culler. 1. Hom and Tensor
A TALE OF TWO FUNCTORS Marc Culler 1. Hom and Tensor It was the best of times, it was the worst of times, it was the age of covariance, it was the age of contravariance, it was the epoch of homology, it
More informationN. CHRISTOPHER PHILLIPS
OPERATOR ALGEBRAS ON L p SPACES WHICH LOOK LIKE C*-ALGEBRAS N. CHRISTOPHER PHILLIPS 1. Introduction This talk is a survey of operator algebras on L p spaces which look like C*- algebras. Its aim is to
More informationPaolo Stellari STABILITY CONDITIONS ON GENERIC K3 SURFACES
Paolo Stellari STABILITY CONDITIONS ON GENERIC K3 SURFACES Joint with D. Huybrechts and E. Macrì math.ag/0608430 Dipartimento di Matematica F. Enriques Università degli Studi di Milano CONTENTS A generic
More informationNotes on p-divisible Groups
Notes on p-divisible Groups March 24, 2006 This is a note for the talk in STAGE in MIT. The content is basically following the paper [T]. 1 Preliminaries and Notations Notation 1.1. Let R be a complete
More informationNonabelian Poincare Duality (Lecture 8)
Nonabelian Poincare Duality (Lecture 8) February 19, 2014 Let M be a compact oriented manifold of dimension n. Then Poincare duality asserts the existence of an isomorphism H (M; A) H n (M; A) for any
More informationON COSTELLO S CONSTRUCTION OF THE WITTEN GENUS: L SPACES AND DG-MANIFOLDS
ON COSTELLO S CONSTRUCTION OF THE WITTEN GENUS: L SPACES AND DG-MANIFOLDS RYAN E GRADY 1. L SPACES An L space is a ringed space with a structure sheaf a sheaf L algebras, where an L algebra is the homotopical
More information1. Algebraic vector bundles. Affine Varieties
0. Brief overview Cycles and bundles are intrinsic invariants of algebraic varieties Close connections going back to Grothendieck Work with quasi-projective varieties over a field k Affine Varieties 1.
More informationarxiv:math/ v1 [math.ra] 20 Feb 2007
ON SERRE DUALITY FOR COMPACT HOMOLOGICALLY SMOOTH DG ALGEBRAS D.SHKLYAROV arxiv:math/0702590v1 [math.ra] 20 Feb 2007 To Leonid L vovich Vaksman on his 55th birthday, with gratitude 1. Introduction Let
More informationIND-COHERENT SHEAVES AND SERRE DUALITY II. 1. Introduction
IND-COHERENT SHEAVES AND SERRE DUALITY II 1. Introduction Let X be a smooth projective variety over a field k of dimension n. Let V be a vector bundle on X. In this case, we have an isomorphism H i (X,
More informationLECTURE X: KOSZUL DUALITY
LECTURE X: KOSZUL DUALITY Fix a prime number p and an integer n > 0, and let S vn denote the -category of v n -periodic spaces. Last semester, we proved the following theorem of Heuts: Theorem 1. The Bousfield-Kuhn
More informationSERRE FINITENESS AND SERRE VANISHING FOR NON-COMMUTATIVE P 1 -BUNDLES ADAM NYMAN
SERRE FINITENESS AND SERRE VANISHING FOR NON-COMMUTATIVE P 1 -BUNDLES ADAM NYMAN Abstract. Suppose X is a smooth projective scheme of finite type over a field K, E is a locally free O X -bimodule of rank
More informationSPECTRAL ENRICHMENTS OF MODEL CATEGORIES
SPECTRAL ENRICHMENTS OF MODEL CATEGORIES DANIEL DUGGER Abstract. We prove that every stable, presentable model category can be enriched in a natural way over symmetric spectra. As a consequence of the
More informationHigher Categories, Homotopy Theory, and Applications
Higher Categories, Homotopy Theory, and Applications Thomas M. Fiore http://www.math.uchicago.edu/~fiore/ Why Homotopy Theory and Higher Categories? Homotopy Theory solves topological and geometric problems
More informationL E C T U R E N O T E S O N H O M O T O P Y T H E O R Y A N D A P P L I C AT I O N S
L A U R E N T I U M A X I M U N I V E R S I T Y O F W I S C O N S I N - M A D I S O N L E C T U R E N O T E S O N H O M O T O P Y T H E O R Y A N D A P P L I C AT I O N S i Contents 1 Basics of Homotopy
More informationMODEL STRUCTURES ON PRO-CATEGORIES
Homology, Homotopy and Applications, vol. 9(1), 2007, pp.367 398 MODEL STRUCTURES ON PRO-CATEGORIES HALVARD FAUSK and DANIEL C. ISAKSEN (communicated by J. Daniel Christensen) Abstract We introduce a notion
More informationFINITE SPECTRA CARY MALKIEWICH
FINITE SPECTRA CARY MALKIEWICH These notes were written in 2014-2015 to help me understand how the different notions of finiteness for spectra are related. I am usually surprised that the basics are not
More informationarxiv:math/ v2 [math.kt] 2 Feb 2007
arxiv:math/0701205v2 [math.kt] 2 Feb 2007 A NEW QUILLEN MODEL FOR THE MORITA HOMOTOPY THEORY OF DG CATEGORIES GONÇALO TABUADA Abstract. We construct a new Quillen model, based on the notions of Drinfeld
More informationCategories and functors
Lecture 1 Categories and functors Definition 1.1 A category A consists of a collection ob(a) (whose elements are called the objects of A) for each A, B ob(a), a collection A(A, B) (whose elements are called
More informationUniversität Regensburg Mathematik
Universität Regensburg Mathematik Brown representability in A 1 -homotopy theory Niko Naumann and Markus Spitzweck Preprint Nr. 18/2009 Brown representability in A 1 -homotopy theory Niko Naumann and Markus
More information9. The Lie group Lie algebra correspondence
9. The Lie group Lie algebra correspondence 9.1. The functor Lie. The fundamental theorems of Lie concern the correspondence G Lie(G). The work of Lie was essentially local and led to the following fundamental
More informationVERY ROUGH NOTES ON SPECTRAL DEFORMATION THEORY
VERY ROUGH NOTES ON SPECTRAL DEFORMATION THEORY 1. Classical Deformation Theory I want to begin with some classical deformation theory, before moving on to the spectral generalizations that constitute
More informationAFFINE PUSHFORWARD AND SMOOTH PULLBACK FOR PERVERSE SHEAVES
AFFINE PUSHFORWARD AND SMOOTH PULLBACK FOR PERVERSE SHEAVES YEHAO ZHOU Conventions In this lecture note, a variety means a separated algebraic variety over complex numbers, and sheaves are C-linear. 1.
More information